README
"A Field Experiment in Motivating Employee Ideas"
by Michael Gibbs, Susanne Neckermann, and Christoph Siemroth, 2016

Note: The analysis uses proprietary data from a private company. The Review of Economics and Statistics is satisfied that the Stata programs and logs of running those programs in this archive comply with the Review's data availability policy (email from October 20, 2014). 

Content of the archive:
- Stata file "tablecreate.do" was used to create the estimates in the tables of the paper, with some manual layout editing afterwards
- The attached self-written ado files are called by the do-file (include in your /ado directory)
- The log file shows the Stata output when running the do-file

Run on Stata 14.1 SE, Windows 7 SP1.

The do file uses several data files:
- person-inactive.dta: Each observation is one author-period with the number of ideas this author submitted aggregated over 13 months (1 period); includes inactive employees (no submitted ideas), includes pretreatment period (period2=0) and treatment period (period2=1)
- person-inactive-expost.dta: Same as person-inactive.dta, except it also includes the post-treatment period
- idealevel.dta: Each observation is one author-idea, i.e., each idea (uniquely identified by ideaid) may have several observations if there were several authors. Each observation has author-specific variables.
- idealevel-expost.dta: Same as idealevel.dta, except it also includes the post-treatment period

Most important dependent variables:
- treatmentper2: dummy, interaction between treatment dummy and period2 dummy (treatment period dummy)
- treatmentper3: dummy, interaction between treatment dummy and period3 dummy (post-treatment period dummy)
- customer1-customer19: dummies indicating the client team
- age/age2: age of employee in years at time of period end or idea submission 
- tenure/tenure2: tenure in years of employee at the company at time of period end or idea submission
- male: gender dummy
- NumAuthors: number of authors submitting the idea
- creative: dummy indicating prior ideator (defined in the paper)
- creativetreatment: interaction of creative dummy with treatment dummy
- creativeperiod2: interaction of creative dummy with treatment period dummy (period2)
- creativetreatmentper2: interaction of creative dummy with treatmentper2 (Creative x Treatment x Period2)
- salaryX: dummy denoting salary group of the employee

Outcome variables:
- shared: dummy, indicates whether idea is shared with client
- implemented: dummy, indicates whether idea is accepted to be implemented or already implemented
- lognetgain: natural logarithm of the net value estimate of the proposed idea
- amountideasnoweight: number of ideas of the employee in the current period (i.e., within 13 months)